Inorg. Chem. 1988, 27, 2592-2595
2592
Contribution from Ames Laboratory and the Department of Chemistry, Iowa State University, Ames, Iowa 5001 1
Reactions of the Superoxochromium(II1) Ion with Transition-Metal Complexes Mark E. Brynildson, Andreja Bakac,* and James H. Espenson* Received October 27, 1987 The reactions of the superoxochromium(II1) ion, CrO*+, with a number of one-electron-reducing agents have been examined in acidic aqueous solutions at 25 OC and 0.10 M ionic strength. The results are consistent with an outer-sphere mechanism for R~(NH~):+ (k = 9.5 x 105 M-1 s-I), Co(sep)2+ (8.5 X lo5 M-' s-I ), and V(H20):' (2.3 X lo5 M-' s-l). The reactions with Fe(H20)62+,Co( [ 14]aneN4)(OH2);+, and Co( [ 151a11eN,)(OH~)~~+ take place by an inner-sphere mechanism characterized by the formation and subsequent decomposition of binuclear intermediates. The rate constants for these reaction$ have values 4.5 X lo3, -7 X lo6, and 6.2 X lo5 M-I s-I, respectively.
Introduction Oxidations t h a t depend on dioxygen are of fundamental importance t o many chemical and biological processes.' Although dioxygen is a n integral part of many redox systems, it is known t o be sluggish in many of its reactions, often utilizing a transition-metal complex to activate it in redox processes. The triplet ground state of dioxygen provides a considerable kinetic barrier t o the autoxidation of normally diamagnetic organic molecules. This barrier is associated with generally very slow reactions involving a change of spin or formation of unstable triplet-state products. This problem can be avoided by the reaction of dioxygen coordinated to a transition metal, where greater spin-orbit coupling considerably reduces t h e kinetic barrier t o a change of spin. In addition, the formation of a metal dioxygen complex may provide sufficient energy to pair the spins.lb Thus the coordination of O2 t o reduced transition metals often yields potent and efficient
oxidants. Little is known about the reactivity patterns of transitionmetal-dioxygen adducts in aqueous solution since the lifetimes of these adducts vary from moderately long t o extremely short. One dioxygen adduct t h a t has been studied in some detail is Co([ 14]aneN4)(OH2)02+.2 The report on the reactivity of this species represents much of what is known about the chemistry of transition-metal-dioxygenadducts. The complex, which readily decomposes in aqueous solutions, is a moderately strong oxidant (estimated Eo 0.3 f 0.1 V).Zb T h e oxidations of several transition-metal complexes with t h e in situ generated Co( [ 141a ~ ~ e N ~ ) ( o H , ) ( oproceed ~ ) ~ + with high rate constants and utilize all three oxidizing equivalents provided by the coordinated superoxide. Another transition-metal-oxygen adduct, Cr022+,3has been characterized r e ~ e n t l y . ~This is a semistable species with a half-life of -15 min in 0.10 M aqueous HC104 at 25 OC under anaerobic conditions. Owing to its relative stability, pronounced spectral features, and known high reactivity toward one reductant (Cr2+), it appeared t o be an interesting candidate for a kinetic and mechanistic study of its reactions with a series of reducing complexes. This paper reports the results of that study. Experimental Section Materials. Dilute aqueous solutions of Cr022' were prepared ac-
-
(a) Neiderhoffer, E. C.; Timmon, J. H.; Martell, A. E. Chem. Rev. 1984,84, 137. (b) Gubelmann, M. H.; Williams, A. F. Struct. Bonding (Berlin) 1983, 55, 1. (c) Spiro, T. G., Ed. Metal Ion Activation of Dioxygen; Wiley: New York, 1980. (d) Hayaishi, O., Ed. Molecular Mechanisms of Oxygen Activation; Academic: New York, 1976. ( e ) Schultz. J.. Cameron. B. F.. Eds. The Molecular Basis for Electron Transport; Academic: New York, 1972. (a) Wong, C.-L.; Endicott, J. F. Inorg. Chem. 1981, 20, 2233. (b) Kumar, K.; Endicott, J. F. Inorg. Chem. 1984, 23, 2447. (c) Munakata, M.; Endicott, J. F. Inorg. Chem. 1984, 23, 3693. (a) Picard, J. Chem. Ber. 1913, 46, 2477. (b) Ardon, M.; Stein, G. J . Chem. SOC.1956, 2095. (c) Kolaczkowski, R. W.; Plane, R. A. Inorg. Chem. 1964, 3, 322. (d) Ilan, Y. A,; Czapski, G.; Ardon, M. Isr. J . Chem. 1975, 13, 15. (e) Sellers, R. M.; Simic, M. G. J . Am. Chem. Soc. 1976, 98, 6145. Brynildson, M. E.; Bakac, A,; Espenson, J. H. J . Am. Chem. SOC.1987, 109, 4579.
0020-1669/88/1327-2592$01.50/0
cording to the literature p r o c e d ~ r e .The ~ ligands5 [ 14]aneN4and [15]aneN4 were obtained from Strem Chemical Co. Samples of Co(mesoMe,[ 14]aneN4)(0H2)2(C104)2 and C o ( t m ~ ) ( O H ~ ) ( C 1were 0 ~ ) ~available in our laboratory reserves, and Co(sep)C13 was a gift from A. M. Sargeson. R U ( N H ~ ) was ~C~ obtained ~ from Aldrich Chemical Co. Solutions of Co( [ 14]aneN4)(OH2)*+ were prepared by mixing a 10% stoichiometricexcess of the ligand with deaerated solutions of C O ( C I O ~ ) ~ Sufficient HC104 was added after the reaction was complete to make the solution 0.10 M in H'. The UV-vis spectrum of anaerobic Co([14]aneN4)(0H2)2+agreed well with the published one (A, = 465 nm, e = 22 M-' cm-1).6 Solutions of Co([15]aneN4)(OH2)2+were prepared similarly and used at the natural pH owing to the low stability of the complex in acidic solutions. The concentrations of the anaerobic stock solutions of Co( [ 15]aneN4)(OH2)22+were determined spectrophotometrically at 497 nm (e = 13.9 M-I cm-').' The concentrations of the stock solutions of Co(meso-Me6[ 14]aneN4)(OH2)22+and Co(tmc)(OH2)2+were determined from the mass of the cobalt complex taken. Solutions of Co(sep)2' and Ru(NH3)?' were prepared by a zinc amalgam reduction of C o ( ~ e p ) ( C l Oand ~ ) ~RU("~)~C~,, respectively, in HzO. Solutions of Fe2+were prepared by a zinc amalgam reduction of acidic aqueous solutions of Fe(C104),. The concentrations of the reduced species was considered to be equal to that of the parent species. The concentration of the C o ( ~ e p ) ~solution + was determined spectrophotometrically at 470 nm (e = 110 M-I cm-I).* The stock solution of vanadium(1V) perchlorate was prepared by ion exchange (Dowex 50W-X8)of the sulfate salt. Vanadium(1V) was eluted with 1 M perchloric acid and standardized spectrophotometrically (A = 760, t = 17.2 M-I ~ m - l ) .The ~ V(H20)2' used in the experiments was prepared by a zinc amalgam reduction of VO(C104)2in 0.10 M HCIO4. Instrumentation. Routine UV-visible spectra were recorded by use of a Perkin-Elmer Lambda Array 3840 UV/VIS spectrophotometer. The kinetic measurements were made with a Durrum stopped-flow spectrophotometer interfaced to an OLIS 3820 data system. Multiwavelength stopped-flow experiments used a George W. Gates Co. GID-25 rapidscan monochromator, which served as the light source when interfaced to the Durrum stopped-flow spectrophotometer. Unless stated otherwise, all the stopped-flow experiments were done under anaerobic conditions at 25 "C and with a cell path length of 2 cm. The single-wavelength kinetic measurements for the outer-sphere electron-transfer reactions were done at one of the two maxima of Cr022+,247 nm (e = 7400 M-I cm-l) and 290 nm (e = 3100 M-I cm-1).3cThe reductants were present in a pseudo-first-order excess relative to [ C r O P ] . The ionic strength was kept constant in all experiments at 0.10 M (HC104 + LiCIO,). The kinetic traces were analyzed by use of the first-order, second-order, and biexponential fitting routines on the OLIS computer system. Results
Reaction between Cr022+and Outer-Sphere-Electron-Transfer Reductants. T h e reactions are first-order in [ C r 0 2 2 + ]and [ R ] (5) [14]aneN4 = 1,4,8,1l-tetraazacyclotetradecane; [15]aneN4 = 1,4,8,12-tetraazacyclopentadecane;tmc = 1,4,8,1l-tetramethyl1,4,8,11-tetraazacyclotetradecane. (6) Heckman, R. A.; Espenson, J. H. Inorg. Chem. 1979, 18, 38. (7) Wong, C.; Switzer, J. A.; Balakrishnan, K. P.; Endicott, J. F. J . Am. Chem. Soc. 1980, 102, 5511. (8) Creaser, I. I.; Harrowfield, J. MacB.; Herlt, A. J.; Sargeson, A. M.; Springborg, J.; Geue, R. J.; Snow, M. R. J . Am. Chem. SOC.1977, 99, 3181. (9) Sykes, A. G.; Green, M. J . Chem. SOC.A 1970, 3221.
0 1988 American Chemical Society
Inorganic Chemistry, Vol. 27, No. 15, 1988 2593
Reactions of CrOZ2+with Transition-Metal Complexes 6.0 W
2
4.5
m 4 LL
0
p
A
3.0
0.00 , 0.05
$
0.10
0.15
0.75
1.50
2.25
3.00
TIME /s
100 0.5,
0
0
1
2
3
1 04[R],,.
4
5
I
6
/M
Figure 1. Dependence of kokd on [R],, ([HClO,] = = 0.10 M, T = 25.0 "C).R = Ru(NH&~+(circles), Co(sep)*+ (squares), V(OH2)62+ (triangles).
( R = RU("~)~'+, Co(sep)2+,and V(HzO)6z+). Chloride ion in the concentration range 3 X lO4-O.05 M has no effect on the reactions of CrOz2+with Ru(NH,):+. It was assumed that the reactions with Co(sep)2+and V2+are also chloride independent. N o spectroscopic evidence for an intermediate was obtained in any of the reactions. The absorbances at 290 and 247 nm decreased during the reaction of Cr022+with RU("~)~'+ and Vz+. The decreasing absorbance-time traces represent directly the consumption of CrO*+ during the reaction, since the other solutes absorb relatively + little compared to CrO*+ at 247 and 290 nm. C ~ ( s e p ) ~absorbs 14000 M-' cm-'), and the absorbances strongly at 247 nm ( e increased in the reaction of CrOzz+with Co(sep)2+. The absorbance change at 247 nm suggests a 1:l stoichiometry for this reaction. The UV spectrum of the products is consistent with the formation of C ~ ( s e p ) ~ + . The results of the kinetic experiments are consistent with an outer-sphere reaction mechanism, eq 1-3. Figure 1 illustrates
-
Cr022+
+R
-
Cr02+
+ R+
-d[CrOZ2+]/dt = kl[Cr022+J[R] kobsd
=
kl [R]
(1) (2) (3)
the three plots of koM vs [R],, suggested by eq 3. A least-squares analysis yielded second-order rate constants k l for the three reagents R u ( N H ~ ) ~ ' +Co(sep)'+, , and V2+ as (9.5 f 0.2) X lo5, (8.5 f 0.2) X lo5,and (2.28 f 0.05) X lo5 M-' s-l, respectively. Reaction between CrOz2+and Fez+. A biexponential reaction is observed with a rapid consumption of Cr022+signaled by the absorbance changes. Once the effect of HCr0,- is allowed for (seenext paragraph), it became evident that the reaction of interest gives rise to a decrease in the absorbance at 290 nm (first stage) and a slower increase in the absorbance at 240 nm (second stage). The consumption of CrOZ2+was monitored at 290 nm, and the formation of Fe3+ at 240 nm (Figure 2). The experiments were done with Fe2+ in pseudo-first-order excess under both aerobic and anaerobic conditions. Oxygen had no effect on the reaction and later experiments utilized air-saturated solutions. In the anaerobic experiments Cr02Z+was degassed for 5 min. This time is sufficient for a buildup of 6 5 X lo4 M HCrO,-., The amount of HCr0,- in solution at a given time depends on the age of the Cr022+solution and initial concentrations of dioxygen and CrO?': The presence of HCr0,- can account for the second of the two reaction stages observed at 290 nm and the
0.0L 0.00 0.25
J
IC
0.50
0.75
"
60
30
90
120
TIME /s
Figure 2. The reaction between C r O F and Fe2+illustrated by a family of absorbance-time profiles at 290 nm (upper figure) and 240 nm (lower figure). All experiments were performed on the same solution at various times over a 2-h period (t = 0 h, A; t = 2 h, B). The first stage at 240 nm and the second stage at 290 nm are caused by the presence of small amounts of HCr04- (see text).
first of the two reaction stages at 240 nm. This assignment, based on a previous study of the HCrO;/Fe(II) system,I0was confirmed by blank experiments with genuine HCr04-. The first stage at 290 nm and the second stage at 240 nm can be assigned to the reaction of CrOzz+with Fez+. The overall absorbance change at 240 nm suggests that one molecule of oxidizes three molecules of Fe2+ to Fe3+ (eq 4). CrO:+
+ 3Fe2+ + 4H+
-
Cr3* + 3Fe3+
+ 2H20
(4)
The reaction initially produces an intermediate wtih an absorption maximum at 320 nm. The increase in absorbance at 320 nm coincides with the loss of Cr02zf at 290 nm (eq 5). The data -d[Cr022+]/dt = d[Int]/dt = kFe[Cr0,2+][Fe2+] ( 5 ) in Figure 3 yield the rate constant kFe= (2.45 f 0.05) X lo4 M-I s-l, independent of [H+] in the range 0.010-0.10 M. There is no absorbance change at 240 nm in this stage. In a subsequent slower reaction the absorbance decreases at 320 nm and increases at 240 nm (Figure 4), according to the rate law of eq 6. The absorbance increase at 240 nm is consistent -d[Int]/dt = k[Int][Fe2+]
(6)
with the formation of two Fe3+ ions per intermediate ion. From the data in Figure 3 one calculates k6 = 60 A 2 M-' s-', independent of [H+] in the range 0.010-0.10 M. A mechanism consistent with all the observations is shown in eq 7-10. The ~
~
~~
~~
(10) Espenson, J. H. J . Am. Chem. SOC.1970, 92, 1880.
2594 Inorganic Chemistry, Vol. 27, No. 15, 1988
+ Fe2+ 5 Cr02Fe4+
Cr02+
(7)
fast
7C r 0 2 H Z ++ Fe3+
(8)
+ Fez+ --gFe3+ + CrOZ++ H 2 0
(9)
Cr02Fe4+ Cr02H2+
Brynildson et al.
kile
+ Fe2+ +
- + - + fast
Fe3+
2H'
net: CrOZ2+ 3Fe2+
4H+
Cr3+
Cr3+
+ H20
(10)
I
v,
3Fe3+
+ 2H20
(4)
binuclear intermediate produced in eq 7 is analogous to Cr02Cr4+ formed in the reaction of CrO?+ and C P e 4 CrOzFe4+is proposed to decompose as in eq 8, owing to the substitutional lability of Fe(III)." Thus, the intermediate observed at 320 nm is assigned to CrO2H2+.I2 The reduction of Cr02H2+by Fe2+ in eq 9 and 10 is consistent with the formation of 2 equiv of Fe3+in the second stage of the reaction. On the basis of the 1:l stoichiometry for the formation of the intermediate (eq 7 and 8) and the absorbance changes in the kinetic experiments, the molar absorptivity of Cr02HZ+is 1300 M-' cm-' at 320 nm and (2-3) X lo3 M-' cm-I at 240 nm. Reactions of Cr022+ with C0([14]aneN~)(OH~)~~+ and Co([ 1S]aneN4)(OHz)2Z+.The UV-vis reaction profiles recorded in the multiwavelength stopped-flow experiments indicated multiphasic kinetics for both reactions. A rapid rise in the absorbance of the reaction mixture in the range 240-400 nm is followed by a slower absorbance decrease in the same wavelength range. The reactions were monitored spectrophotometrically in single-wavelength stopped-flow experiments a t 320 nm, the wavelength of the maximum absorbance change. The first stage for the reaction of CrOz2+with Co([14]aneN4)(OHz)~+ is very fast. Efforts to lower the observed rate by lowering the concentration of the excess reagent, Co( [ 14]aneN4)(OH2);+, were not feasible owing to the experimental uncertainty associated with the handling of the extremely oxygen-sensitive Co( [ 14]aneN4)(OH2)22+at the M) concentration levels. A single kinetic meavery low ( surement at 25 O C yielded the rate constant kco[l4] 7 X lo6 M-l s-l. The second stage has an observed rate constant of 2.5 s-l, and the third, a rate constant of 0.064 s-'. It is possible that some of the absorbance changes in the Co[n]aneN4)(Hz0),2+ reactions are caused by trace amounts of chromate, like the reaction with Fe2+. The first stage for the reaction of Co( [ 15]aneN4)(OH2)22+is first-order in [CrOZ2+]and first-order in [Co( [ 15]aneN4)(OH2)22+]. Figure 3 illustrates a plot of kow vs [Co([15]aneN4)(0H2)2+],. A least-squares analysis yields a rate constant k,[151 = (6.2 f 0.3) X lo5 M-l s-l. The second stage is indeand [H+]. An pendent of [Cr022+],[C0([15]aneN~)(OH,)~~+], average value for the first-order rate constant for the second stage is 0.48 f 0.02 s-l. The data for both cobalt(I1) complexes are consistent with an inner-sphere mechanism, eq 11 and 12, analogous to that proposed
\ n
1
1 03[Fe( I
/M
Figure 3. Dependence of koM on [Fe2+IPV (first stage, circles; second ],~ (25 OC, stage, squares) and of ka on [ C 0 ( [ 1 5 ] a n e N ~ ) ~ +(triangles) [H30+]= w = 0.10 M). The data symbolized by squares were multiplied by lo2. 1 .o
t
-
+
kfit.1
Cr022+ Co( [ n ] a r ~ e N ~ ) ~ +
Cr02Co([r~]aneN,)~+ (1 1)
n = 14, 15
-
Cr02Co([ r ~ ] a n e N ~ ) ~ products +
(12)
for the reaction with Fe2+. The binuclear intermediates produced in these reactions are not expected to dissociate as readily as Cr02Fe4+owing to the substitutional inertness of Co(II1). The absorption maximum at 320 nm is thus assigned to the binuclear complexes. The similarity in the spectral features between Cr02H2+and [CrO,Co( [n]aneN4)I4+ seems reasonable, since (1 1) Basolo, F.; Pearson, R. G. Mechanisms oflnorganic Reactions; Wiley: New York, 1967. (12) The protonated form of the peroxochromium complex, Cr02H2+,is probably dominant under our experimental conditions, like the related cobalt complex (H20)([14]aneN4)Co02H2+.2b
200
250
300
350
400
WAVELENGTH (nm)
Figure 4. Absorbance-wavelength profiles for the loss of the intermediate formed in the reaction of CrO?+ and Fez'. The absorbance decreases at 320 nm ( k = 0.012 s-I) and increases at 240 nm ( k = 0.011 s-l). The total collection time is 430 s ([H30+] = p = 0.10 M, [ C r 0 P l o = 3.2 X M, [Fe2+Io= 2.5 X lo4 M).
Cr02+is the chromophore in both species. Other Reactions of CrOZ2+. The reaction with Co(mesoMe6[14]aneN4)2+did not conform to the general pattern observed with other reductants studied. No reaction was detected between Cr02,+ and Co(tmc)2+, apart from the rapid decomposition of Co(tmc),+ in acidic medium. Discussion Reactions with V(HzO)62+,Ru("~)~'+, and Co(sep)'+. An outer-sphere electron transfer seems to be the most reasonable mechanistic assignment for the reactions of CrOZ2+with substitutionally inert V(H20)s2+,R U ( N H ~ ) ~ 'and + , Co(sep)'+. The treatment of the data according to the Marcus theory is not feasible since neither the reduction potential nor the self-exchange rate constant for CrO?+/+ is known. The similarity in the observed rate constants for the three reductants, as well as the mild trend > kRU(NH,)62+ > kVz+),has been well within the series (kCo(scp)z+ established in other outer-sphere reaction^,^^*'^-^^ supporting the (13) Ramasami, T.; Endicott, J. F. J . Am. Chem. SOC.1985, 107, 389. (14) Candlin, J. P.; Halpern, J.; Trimm, D. L. J . Am. Chem. SOC.1964,86, 1019. (15) Endicott, J. F.; Taube, H. J . Am. Chem. Soc. 1968, 86, 1686.
Reactions of Cr022+with Transition-Metal Complexes Table I. Summary of Kinetic Data for Reductions of CrO?' and Co( [ 14]aneN4)(OH2)022' oxidant klM-' s-I Ef, V Co([14]aneN4)reductant (vs NHE) (H20)5Cr01t" (OH2)02" RU("&*+ 0.06~ (9.5f 0.2) x 105 (2.3 0.1) x 105" Co(sep)Z+ -0.30d (8.5 0.3)X 10' -lo6' V(OH2)62+ -0.226' (2.3f 0.1)X lo5 (1.8 0.3) X lo5' C0([14]aneN~)~' 0.4211 -7 X lo6 (4.9 0.4) x 105' C0([15]aneN~)~+ 0.65b (6.2f 0.3) X los (3.7f 0.4) X lo4' Fe(OH2):' 0.749 (4.5 0.2) x 103 (1.1 0.1) x 10" cr(OH1)6~' -0.418 -8 X losh
*
* **
*
*
"[HCIO,] = = 0.10 M. bReference 2b. CReference 12. dSargeson, A. M. Chem. Ber. 1979, 15, 23. eLatimer, W.M. Oxidation Potentials, 2nd ed.; Prentice-Hall: Englewood Cliffs, NJ, 1952. fEndicott, J. F.; Durham, B.; Glick, M. D.; Anderson, T. J.; Kuszaj, J. M.; Schmonsees, W. G.; Balakrishnan, K. P. J . A m . Chem. SOC.1981, 103, 1431. SYee, E. L.; Cave, R. J.; Guyer, K. L.; Tyma, P. D.; Weaver, M. J. J. Am. Chem. SOC. 1979, ZOZ, 1131. *Reference 4. '[HC104] = 0.1 M, [NaC104] = 0.2 M. '[HCIO,] + [NaC104] = 0.1 M.
assignment in the present system. In none of the reactions was there any evidence obtained for the reverse of eq 1. All the reactions obeyed pseudo-first-order kinetics cleanly, and no appreciable intercepts are noted in Figures 1 and 3. Thus the reduction potential for C r 0 2 + / + in 0.10 M HC104must be at least as high as the reduction potential of the mildest reductant, R U ( N H & ~ +(0.06 V).22 This is consistent with the estimated Eo for Co([14]aneN4)02+/+of 0.3 V,2bsince the two oxygen adducts react at similar rates with all the complexes studied (Table I). Reactions with C ~ ( [ n ] a n e N ~ )(n ~ += 14,15) and Fe2+. Since these reducing agents feature ligand substitution rate constants (16) Przystas, T.J.; Sutin, N. J . Am. Chem. SOC.1973, 95, 5545. (17) Endicott, J. F.;Brubaker, G.R.; Ramasami, T.; Kumar, K.; Dwarakanath, K.; Cassel, J.; Johnson, D. Inorg. Chem. 1983, 22, 3754. (18) Davies, R.; Green,M.; Sykes, A. G.J. Chem. SOC.Dalton Trans. 1972, 1171. (19) Chou, M.; Creutz, C.; Sutin, N. J . Am. Chem. SOC.1977, 99, 5615. (20) Jacks, C. A.; Bennett, L. E. Inorg. Chem. 1974, 13, 2035. (21) Marcus, R. A,; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265. (22) Lim, H.S.;Barclay, D. J.; Anson, F. Inorg. Chem. 1972, !I, 1460.
Inorganic Chemistry, Vol. 27, No. 15, 1988 2595 in excess of the observed rate constants for reduction of CrO?+, they can react by inner-sphere pathways. The direct observation of intermediates in all three reactions strongly indicates the formation of p-peroxo c ~ m p l e x e s . ~ ~This - ~ ~ conclusion is strengthened by the similarity in the chemistry of C r 0 2 + and Co( [ 14]aneN4)022+.Detailed product analysis of the reactions of the latter complex with several reductants has shown that these reactions form metastable p-peroxo complexes. The formation of the p-peroxo adducts makes a large contribution to the overall free energy change, making adduct formation (inner-sphere pathway) the preferred reaction p a t h ~ a y . ~This ~ , is ~ also ~ expected to be true for the inner-sphere reactions of The intermediates observed in the reactions of CrO2+ with all three reductants exhibit absorption maxima at 320 nm. A species with a similar spectrum has been observed in a previous study of the reaction of Cr022+with N2H5+.25 It was presumed to be a feature of the subsequent chromium chemistry and not directly related to the reaction of interest. A similar spectrum is observed when Cr(OH2):+ and H202are allowed to react for a long time. Given the complexity of the chemistry involved26in all of these reactions, it is likely that some common intermediates are formed. The decomposition of the unstable intermediates, Cr02H2+ (formed in the reaction with Fe2+)and Cr02Co([ n ] a ~ ~ e N ~take )~+, place by different mechanisms. The sterically hindered bimetallic complexes decompose in unimolecular processes, whereas Cr02H2+ shows a typical reactivity of a peroxide toward Fe2+. Owing to the very low concentrations of in solution, the detailed product analysis was not feasible. Acknowledgment. This research was supported by the National Science Foundation, Grant CHE-8418084. Some of the facilities for this work were provided by Ames Laboratory, which is operated by Iowa State University under Contract W-7405-Eng-82. We are grateful to Dr. A. Sargeson for a gift of Co(sep)Cl,. (23) A referee has suggested a possibility that CrOZ2+is a rapidly equilibrated mixture of Cr111(02-)2+ and Cr'V(0~-)2C, with only the former yielding the binuclear products. We cannot completely rule out this possibility, although an analysis4 of a series of stability constants for chromium(II1) complexes has suggested that CrO;+ is best described as Cr111(02-)2+. (24) Endicott, J. F.;Kumar, K. ACS Symp. Ser. 1982, No. 198, 425. (25) Bruhn, S.L.; Bakac, A,; Espenson, J. H. Inorg. Chem. 1986, 25, 535. (26) Adams, A. C.; Crook, J. R.; Bockhoff, F.; King, E. L. J . Am. Chem. SOC.1968, 90, 5761.
